IRMA-International.org: Creator of Knowledge
Information Resources Management Association
Advancing the Concepts & Practices of Information Resources Management in Modern Organizations

Automated Detection of Cardiovascular Arrhythmias Using Machine Learning and Deep Learning on ECG Signals

Automated Detection of Cardiovascular Arrhythmias Using Machine Learning and Deep Learning on ECG Signals
View Sample PDF
Author(s): Akash Saxena (Compucom Institute of Technology and Management, Jaipur, India), Satish Kumar Alaria (Government of Rajasthan, India), Ashish Raj (Poornima University, India), Apoorva Sharma (Poornima University, India)and Arpita Sharma (Swami Keshvanand Institute of Technology, Management, and Gramothan, Jaipur, India)
Copyright: 2026
Pages: 38
Source title: The Emerging Role of Advanced Technologies in Neurological Diseases
Source Author(s)/Editor(s): Adnène Arbi (National Institute of Applied Sciences and Technology, University of Carthage, Tunisia & Laboratory of Mathematical Engineering, Tunisia Polytechnic School, University of Carthage, Tunisia)and Walid Ben Ameur (Faculty of Sciences of Gabes, Tunisia)
DOI: 10.4018/979-8-3373-2706-8.ch003

Purchase

View Automated Detection of Cardiovascular Arrhythmias Using Machine Learning and Deep Learning on ECG Signals on the publisher's website for pricing and purchasing information.

Abstract

This study explores various machine learning and deep learning approaches to classify normal and arrhythmic heartbeats using the MIT-BIH Arrhythmia dataset. Feature extraction techniques, including Discrete Fourier Transform, Discrete Wavelet Transform, and Continuous Wavelet Transform, were employed to enhance signal representation. Machine learning classifiers were tested, along with deep learning-based models and a custom Convolutional Neural Network (CNN) trained on CWT-generated scalograms. The results indicate that deep learning models significantly outperform traditional machine learning classifiers, with the custom CNN achieving the highest accuracy of 82.76% when trained on augmented CWT scalogram images. While KNN (97.76%) and SVM (98%) performed well, their dependency on feature engineering and computational limitations restricted their scalability. This study underscores the potential of AI-driven ECG classification to revolutionize automated arrhythmia detection, enabling real-time monitoring and early intervention in clinical settings.

Related Content

V. Leela, R. Sangeetha, S. Geetha, B. Deepa. © 2026. 38 pages.
A Prabhu Chakkaravarthy, Dhanalakshmi Jaganathan. © 2026. 20 pages.
Hasini Balage, Darshana Sedera. © 2026. 24 pages.
Dilek Gümüş. © 2026. 34 pages.
Fawaz Azizieh, Bulent Yilmaz. © 2026. 46 pages.
Kutay Icoz. © 2026. 54 pages.
Rajganesh Nagarajan, G. Kavitha. © 2026. 36 pages.
Body Bottom